Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A network switch comprising: a plurality of ports, the network switch being configured to switch packets of a layer 2 network received on the plurality of ports; a memory for storing a tunneling engine computer program; and a processor configured to execute the tunneling engine, wherein the processor is configured to identify a second switch configured to switch packets of the layer 2 network, the identification including detecting that the second switch is connected to the network switch over a layer 3 connection, wherein the tunneling engine is configured to create a tunnel over the layer 3 connection between the network switch and the second switch that is maintained in said memory in a global MAC address table to exchange packets of the layer 2 network, the global MAC address table of said network switch and said second switch maintained with updates of information regarding the tunnel and other tunnels, the exchange of packets over the tunnel including encapsulation and decapsulation of the packets that are exchanged between the network switch and the second switch, wherein when the processor determines that a received packet of the layer 2 network is from a first node and addressed to a second node connected to the second switch, the processor creates an encapsulation flow on the network switch to encapsulate packets from the first node to the second node over the tunnel that was accessed from said memory for said exchange of packets; wherein the tunnel is created without processing by the first node or the second node, wherein the processor is configured to exchange data between the network switch and the second switch to maintain the updates of information regarding the tunnel and other tunnels stored or to be stored in the global MAC address table of the memory.
Network switching technology for efficiently forwarding Layer 2 network packets between network devices. The problem addressed is enabling communication between Layer 2 devices that are not directly connected and instead communicate over a Layer 3 infrastructure. This invention describes a network switch with multiple ports for receiving Layer 2 packets. It includes a memory containing a tunneling engine program and a processor that executes this engine. The processor identifies another network switch capable of Layer 2 packet switching by detecting a Layer 3 connection between the two switches. A tunnel is established over this Layer 3 connection between the network switch and the identified second switch. This tunnel information, along with details of other tunnels, is stored in a global MAC address table within the network switch's memory. This table is continuously updated through communication with the second switch. When a Layer 2 packet arrives destined for a node connected to the second switch, the network switch encapsulates the packet. This encapsulation, along with subsequent decapsulation at the destination, occurs over the established tunnel without requiring any processing from the originating or destination nodes. The network switch actively exchanges data with the second switch to keep the global MAC address table, which stores tunnel information, up-to-date.
2. The network switch of claim 1 , further including a switch fabric, wherein the switch fabric is configurable by the processor to perform tunneling to the second switch for one or more flows.
A network switch includes a processor and a switch fabric. The switch fabric is configurable by the processor to perform tunneling to a second switch for one or more data flows. This configuration allows the switch to encapsulate and forward traffic between network segments while maintaining flow isolation. The tunneling capability enables the switch to support virtualized network environments, such as overlay networks, by extending Layer 2 connectivity across Layer 3 boundaries. The processor dynamically manages the switch fabric to establish and terminate tunnels based on network conditions or administrative policies. This approach improves scalability and flexibility in network architectures by enabling efficient traffic forwarding across distributed network domains. The solution addresses challenges in modern data center and cloud environments where seamless connectivity between virtualized endpoints is required. The switch fabric's programmability ensures compatibility with various tunneling protocols, such as VXLAN or GRE, allowing integration into diverse network infrastructures. The system enhances network performance by reducing latency and optimizing bandwidth utilization through intelligent flow-based tunneling.
3. The network switch of claim 2 , wherein the switch fabric is connected to the processor via PCIExpress.
A network switch includes a switch fabric and a processor, where the switch fabric is connected to the processor via a PCI Express interface. The switch fabric is configured to route data packets between multiple ports, and the processor is configured to manage the switch fabric, including configuring routing tables, monitoring traffic, and handling control plane functions. The PCI Express connection provides high-speed data transfer between the switch fabric and the processor, enabling efficient communication for both data plane and control plane operations. The switch may also include additional features such as quality of service (QoS) management, traffic shaping, and security functions like packet filtering and encryption. The design ensures low-latency, high-throughput packet forwarding while maintaining flexibility in network management. This architecture is particularly useful in data centers, enterprise networks, and high-performance computing environments where fast, reliable, and scalable network switching is required. The PCI Express connection ensures compatibility with modern computing systems, allowing seamless integration with servers and other network devices.
4. The network switch of claim 3 , wherein the switch fabric is connected to the processor via ethernet utilizing openflow protocol.
A network switch includes a switch fabric and a processor. The switch fabric is connected to the processor via Ethernet using the OpenFlow protocol. The switch fabric is configured to receive and forward data packets between multiple network ports. The processor is configured to manage and control the switch fabric, including configuring forwarding rules, monitoring traffic, and enforcing network policies. The OpenFlow protocol enables the processor to programmatically control the switch fabric by installing and modifying flow rules, allowing for dynamic and centralized network management. This setup enhances network flexibility, scalability, and security by enabling software-defined networking (SDN) capabilities, where the control plane is decoupled from the data plane. The switch fabric may also include additional features such as quality of service (QoS) enforcement, traffic shaping, and port mirroring to optimize network performance. The Ethernet connection ensures high-speed communication between the switch fabric and the processor, while the OpenFlow protocol provides a standardized interface for controlling the switch fabric's behavior. This design allows for efficient traffic management, improved network visibility, and centralized control over network operations.
5. The network switch of claim 1 , wherein the network switch is configured to exchange information with other switches supporting a common network device operating system (ndOS) to share data from the global MAC address table for nodes coupled to any of the switches supporting ndOS.
A network switch is designed to facilitate communication within a network by managing and forwarding data packets between connected devices. A key challenge in network management is efficiently maintaining and sharing MAC address tables across multiple switches to ensure seamless communication and reduce broadcast traffic. This invention addresses this problem by enabling a network switch to exchange information with other switches that support a common network device operating system (ndOS). The switch is configured to share data from its global MAC address table, which contains MAC addresses of nodes connected to any of the switches running ndOS. This shared data allows switches to update their local MAC address tables dynamically, improving network efficiency and reducing unnecessary broadcast traffic. The global MAC address table ensures that switches have up-to-date information about device locations, enabling faster and more accurate packet forwarding. The shared operating system (ndOS) ensures compatibility and interoperability between switches, allowing them to exchange MAC address information seamlessly. This approach enhances network performance by minimizing broadcast storms and improving traffic management across the network.
6. The network switch of claim 1 , wherein when an ARP message or a MAC miss in a MAC table associated with the layer 2 network is received by one of the ndOS switches, the ndOS switches share information regarding port, switch, MAC address, and VLAN based on the ARP message or MAC miss.
A network switch system is designed to improve address resolution and MAC address learning in layer 2 networks. The problem addressed is the inefficiency in traditional networks where switches operate independently, leading to delays in ARP resolution and MAC address table updates. This system includes multiple ndOS switches that dynamically share information when an ARP message or a MAC miss occurs. Upon receiving an ARP message or detecting a MAC address not found in the local MAC table, the switches exchange details such as the port identifier, switch identifier, MAC address, and VLAN information. This shared data allows the switches to update their MAC tables more quickly and accurately, reducing broadcast traffic and improving network performance. The system ensures that all switches in the network have synchronized MAC address mappings, enhancing efficiency in packet forwarding and reducing latency. The solution is particularly useful in large-scale networks where rapid address resolution and MAC learning are critical for optimal performance.
7. The network switch of claim 1 , wherein the first node and the second node have IP addresses in a same subnet associated with the layer 2 network, wherein the first node and the second node are one of a server, or a mobile device, or a virtual machine, or a personal computing device.
A network switch is designed to facilitate communication between nodes in a layer 2 network, particularly when the nodes share the same subnet IP addresses. The switch includes a first node and a second node, each capable of being a server, mobile device, virtual machine, or personal computing device. The nodes are configured to communicate directly over the layer 2 network without requiring additional routing or gateway devices, as they are part of the same subnet. The switch ensures efficient data transmission by managing layer 2 connectivity while maintaining compatibility with standard IP addressing schemes. This design simplifies network architecture by eliminating the need for complex routing configurations when nodes are within the same subnet, improving performance and reducing latency. The solution is particularly useful in environments where multiple devices, such as servers, virtual machines, and personal computing devices, must communicate seamlessly over a local network. The switch's ability to handle diverse node types ensures broad applicability across different network setups.
8. The network switch of claim 1 , wherein tunneling messages between the network switch and the second switch is transparent to the first node and to the second node.
A network switch is designed to facilitate communication between a first node and a second node, where the first node is connected to the network switch and the second node is connected to a second switch. The network switch includes a processor and a memory storing instructions that, when executed by the processor, cause the network switch to establish a tunnel between itself and the second switch. This tunnel enables the transmission of messages between the network switch and the second switch while ensuring that the tunneling process remains transparent to both the first node and the second node. The transparency means that the nodes are unaware of the tunneling mechanism, allowing them to communicate as if they were directly connected without any intermediate processing or modification of the messages. The network switch may also include a plurality of ports for connecting to multiple nodes and switches, and it may support various tunneling protocols to facilitate secure and efficient message transmission. The system ensures seamless communication while maintaining the integrity and transparency of the data exchange between the nodes.
9. The network switch of claim 1 , wherein the processor identifies the second switch when the processor receives an administrator command to connect fabrics of the network switch and the second switch.
A network switch is designed to manage and optimize data traffic in a network environment. A common challenge in network management is efficiently connecting multiple network switches to form a unified fabric, allowing seamless communication between different network segments. This requires precise identification and coordination between switches to ensure proper data routing and avoid conflicts. The network switch includes a processor that facilitates the connection of multiple switches into a unified fabric. When an administrator issues a command to connect the fabrics of the network switch and a second switch, the processor identifies the second switch. This identification process ensures that the switches can establish a proper connection, enabling data to flow between them without disruptions. The processor may use various methods, such as querying network topology data or detecting specific signals from the second switch, to accurately identify it. Once identified, the switches can synchronize their configurations, update routing tables, and establish secure communication channels, allowing the network to operate as a single, cohesive unit. This improves network efficiency, reduces latency, and enhances overall performance by eliminating bottlenecks and ensuring smooth data transmission across the connected switches.
10. The network switch of claim 1 , wherein the processor identifies the second switch when the processor receives a fabric connect message from the second switch.
A network switch is designed to operate within a network fabric, where multiple switches are interconnected to form a scalable and resilient network infrastructure. A key challenge in such environments is efficiently identifying and managing connections between switches to ensure seamless communication and data routing. This invention addresses this problem by enhancing a network switch with a processor that can dynamically detect and establish connections with other switches in the fabric. The network switch includes a processor configured to identify a second switch when it receives a fabric connect message from that second switch. The fabric connect message is a signaling mechanism used to establish communication between switches in the network fabric. Upon receiving this message, the processor recognizes the second switch as a valid peer within the fabric, enabling the switches to exchange routing information, synchronize configurations, and maintain network topology awareness. This dynamic identification process ensures that the network fabric can scale efficiently, automatically incorporating new switches as they join the network without manual intervention. The processor may also be responsible for other functions, such as managing traffic flow, enforcing security policies, and monitoring link statuses. By integrating the fabric connect message detection into the switch's operations, the invention simplifies network deployment and reduces administrative overhead, making it particularly useful in large-scale data center or enterprise environments where network agility and reliability are critical.
11. The network switch of claim 1 , wherein the tunnel is transparent to the first node and the second node.
The network switch creates a tunnel that allows the two connected devices to communicate as if the tunnel wasn't there.
12. A method executed by a network switch having a memory and a processor, comprising: identifying, at the network switch configured to switch packets of a layer 2 network, a second switch configured to switch packets of the layer 2 network, the identifying including detecting that the second switch is connected to the network switch over a layer 3 connection; creating, at the network switch, a tunnel over the layer 3 connection between the network switch and the second switch that is maintained in the memory in a global MAC address table to exchange packets of the layer 2 network, the global MAC address table of said network switch and said second switch maintained with updates of information regarding the tunnel that was created and other created tunnels, the exchange of packets over the tunnel including encapsulation and decapsulation of the packets that are exchanged between the network switch and the second switch; receiving, at the network switch, a packet of the layer 2 network, the packet being from a first node and addressed to a second node; determining, at the network switch, that the second node is connected to the second switch; detecting that the network switch is connected to the second switch via a tunnel implemented over a layer 3 connection that was accessed from said memory for said exchange of packets; and creating, on the network switch, a first encapsulation flow to encapsulate packets from the first node to the second node over the tunnel and a first decapsulation flow to decapsulate packets from the second node to the first node, wherein the second switch creates a second encapsulation flow to encapsulate packets from the second node to the first node over the tunnel and a second decapsulation flow to decapsulate packets from the first node to the second node; wherein the tunnel is created without processing by the first node or the second node, wherein the processor is configured to exchange data between the network switch and the second switch to maintain the updates of information regarding the tunnel and other tunnels stored or to be stored in the global MAC address table of the memory.
This invention relates to network switching in layer 2 networks, specifically addressing the challenge of maintaining connectivity across layer 3 connections. In traditional layer 2 networks, switches rely on direct layer 2 adjacency to forward packets, but when switches are connected over a layer 3 network, standard layer 2 forwarding fails. The invention solves this by enabling network switches to dynamically create and manage tunnels over layer 3 connections, allowing seamless packet exchange between switches as if they were directly connected in a layer 2 domain. A network switch identifies another switch connected via a layer 3 link and establishes a tunnel between them, storing tunnel information in a global MAC address table shared between the switches. The tunnel supports bidirectional packet exchange through encapsulation and decapsulation, with the switches dynamically updating the global MAC table to reflect tunnel status and other tunnels. When a packet is received from a source node destined for a node connected to the second switch, the network switch detects the tunnel and creates encapsulation and decapsulation flows for the traffic. The second switch similarly creates corresponding flows for return traffic. The tunnel operates transparently to the end nodes, with switches exchanging data to maintain the global MAC table. This approach extends layer 2 connectivity over layer 3 networks without requiring modifications to the end nodes.
13. The method as recited in claim 12 , further including: sending the packet from the network switch to the second switch over the tunnel using the encapsulation and decapsulation, the tunnel being created without processing by the first node or the second node.
A method for packet transmission in a network involves sending a packet from a network switch to a second switch over a tunnel. The tunnel is established without requiring processing by either the first node or the second node, which are endpoints of the tunnel. The packet is encapsulated before transmission and decapsulated upon receipt, ensuring secure and efficient data transfer. This method is particularly useful in network environments where direct processing by nodes is undesirable or unnecessary, improving efficiency and reducing latency. The encapsulation and decapsulation processes ensure that the packet remains intact and secure during transmission. The tunnel creation is automated, eliminating the need for manual configuration or intervention by the nodes, which simplifies network management and enhances scalability. This approach is beneficial in scenarios where low-latency, high-throughput communication is required, such as in data centers or cloud computing environments. The method ensures reliable packet delivery while minimizing overhead and complexity.
14. The method as recited in claim 12 , wherein the first node is a first virtual machine (VM) in a first server and the second node is a second VM in a second server.
15. The method as recited in claim 14 , wherein the tunnel between the network switch and the second switch is transparent to the first server and the second server, wherein the tunnel allows the first server and the second server to avoid having a direct tunnel between the first server and the second server which would require resource utilization in the first server and the second server.
This invention relates to network communication systems, specifically methods for optimizing data transfer between servers using intermediate switches. The problem addressed is the inefficiency of direct server-to-server tunnels, which consume significant processing and memory resources on the servers. The solution involves establishing a tunnel between a network switch and a second switch, which acts as an intermediary for communication between a first server and a second server. This tunnel is transparent to both servers, meaning they operate as if communicating directly while avoiding the need for a direct tunnel. By offloading the tunneling process to the switches, the servers conserve resources that would otherwise be used for managing the tunnel. The method ensures seamless data transfer without requiring servers to establish and maintain direct tunnels, reducing their computational load and improving overall network efficiency. The switches handle the encapsulation and decapsulation of data packets, allowing the servers to focus on their primary functions. This approach is particularly useful in data centers and cloud environments where multiple servers frequently exchange data.
16. The method as recited in claim 12 , wherein sending the packet over the tunnel further includes: encapsulating the packet into a VXLAN message.
A method for optimizing network packet transmission involves sending packets over a virtualized network tunnel to improve efficiency and security. The method addresses challenges in modern data networks, such as scalability, isolation, and compatibility with legacy systems, by leveraging virtual extensible LAN (VXLAN) encapsulation. VXLAN is a network virtualization technology that extends Layer 2 networks over Layer 3 infrastructure, enabling seamless communication across distributed environments. The method includes encapsulating packets into VXLAN messages before transmission. This encapsulation process involves adding a VXLAN header to the original packet, which includes metadata such as a VXLAN Network Identifier (VNI) to distinguish between different virtual networks. The encapsulated packet is then transmitted over the tunnel, ensuring secure and efficient delivery. This approach allows for the creation of isolated network segments within a shared physical infrastructure, improving traffic management and reducing congestion. The method also supports interoperability with existing network protocols and devices, ensuring compatibility with legacy systems while enhancing performance. By using VXLAN encapsulation, the method enables dynamic network segmentation, simplifying network management and reducing operational overhead. This solution is particularly useful in data center environments, cloud computing, and enterprise networks where scalability and isolation are critical requirements.
17. The method as recited in claim 12 , wherein the network switch and the second switch are not directly connected over a layer 2 connection when the tunnel is created, wherein the network switch and the second switch exchange information to share data from the global MAC address table for nodes coupled to the network switch and the second switch, wherein detecting that the network switch is connected to the second switch is based on information in the global MAC address table.
This invention relates to network switching and specifically to methods for managing MAC address tables in network environments where switches are not directly connected via a layer 2 connection. The problem addressed is the inefficiency in MAC address learning and forwarding when switches lack direct layer 2 connectivity, leading to potential network performance degradation or misrouting. The method involves a network switch and a second switch that are not directly connected over a layer 2 connection when a tunnel is created between them. Despite this lack of direct connectivity, the switches exchange information to share data from their global MAC address tables, which contain entries for nodes coupled to each switch. This sharing allows the switches to maintain an updated global view of MAC addresses across the network. The detection of the connection between the network switch and the second switch is based on information present in the global MAC address table, ensuring accurate and dynamic network topology awareness. This approach improves MAC address learning and forwarding efficiency, particularly in scenarios where direct layer 2 connectivity is unavailable.
18. A non-transitory computer-readable storage medium storing a computer program to be executed by a network switch having a memory and a processor, the computer-readable storage medium comprising: program instructions for identifying, at the network switch configured to switch packets of a layer 2 network, a second switch configured to switch packets of the layer 2 network, the identifying including detecting that the second switch is connected to the network switch over a layer 3 connection; program instructions for creating, at the network switch, a tunnel over the layer 3 connection between the network switch and the second switch that is maintained in the memory in a global MAC address table to exchange packets of the layer 2 network, the global MAC address table of said network switch and said second switch maintained with updates of information regarding the tunnel that was created and other created tunnels, the exchange of packets over the tunnel including encapsulation and decapsulation of the packets that are exchanged between the network switch and the second switch; program instructions for receiving, at the network switch, a packet of the layer 2 network, the packet being from a first node and addressed to a second node; program instructions for determining, at the network switch, that the second node is connected to the second switch; and program instructions for creating an encapsulation flow on the network switch to encapsulate packets from the first node to the second node over the tunnel; wherein the tunnel is created without processing by the first node or the second node, wherein the processor is configured to exchange data between the network switch and the second switch to maintain the updates of information regarding the tunnel and other tunnels stored or to be stored in the global MAC address table of the memory.
This invention relates to network switching in a layer 2 network, specifically addressing the challenge of maintaining connectivity across layer 3 connections without requiring end-node participation. The solution involves a network switch that identifies another switch connected via a layer 3 link, then establishes a tunnel between them to facilitate layer 2 packet exchange. The tunnel is maintained in a global MAC address table shared between the switches, which is updated dynamically to reflect tunnel status and other connections. The switch receives packets from a source node destined for a destination node, determines if the destination is reachable via the second switch, and creates an encapsulation flow to route packets through the tunnel. The tunnel operates transparently, requiring no configuration or processing by the end nodes. The switches exchange data to synchronize the global MAC address table, ensuring accurate routing decisions. This approach enables seamless layer 2 communication over layer 3 infrastructure, improving network flexibility and scalability without modifying existing nodes.
19. The storage medium as recited in claim 18 , wherein the packet is addressed to an IP address associated with the layer 2 network.
A system and method for network packet processing involves a storage medium storing instructions that, when executed, cause a processor to analyze network traffic. The system intercepts packets transmitted over a network, examines their contents, and determines whether they should be forwarded or blocked based on predefined rules. The system includes a packet inspection module that identifies packets containing specific data patterns or meeting certain criteria, such as being addressed to a particular IP address. The system also includes a filtering module that applies rules to determine whether to allow or deny the transmission of these packets. The system is designed to operate within a layer 2 network, where packets are addressed to an IP address associated with that network. The system may also include a logging module to record packet transmission events for monitoring and analysis. The system is particularly useful for enhancing network security by preventing unauthorized or malicious traffic from reaching its destination. The system can be implemented in hardware, software, or a combination of both, and may be integrated into existing network infrastructure or deployed as a standalone device. The system is designed to operate efficiently with minimal impact on network performance while providing robust security features.
20. The storage medium as recited in claim 18 , wherein tunneling of messages between the network switch and the second switch is transparent to the first node and the second node.
The invention relates to a storage medium containing instructions for managing message tunneling in a network environment involving multiple switches and nodes. The system includes a network switch connected to a first node and a second switch, which is in turn connected to a second node. The storage medium stores instructions that, when executed, enable the network switch to receive a message from the first node, determine that the message is intended for the second node, and forward the message to the second switch. The second switch then delivers the message to the second node. The tunneling process is designed to be transparent to both the first and second nodes, meaning the nodes operate without awareness of the intermediate tunneling steps. This transparency ensures seamless communication while maintaining compatibility with existing network protocols. The system may also include additional features such as error handling, message prioritization, or encryption to enhance reliability and security. The invention addresses the challenge of efficiently routing messages in complex network topologies while minimizing overhead and ensuring compatibility with legacy systems.
21. The storage medium as recited in claim 18 , wherein the first node is a first virtual machine (VM) in a first server and the second node is a second VM in a second server, wherein the tunnel between the network switch and the second switch is transparent to the first server and the second server, wherein the tunnel allows the first server and the second server to avoid having a direct tunnel between the first server and the second server which would require resource utilization in the first server and the second server.
This invention relates to virtual machine (VM) communication in a distributed computing environment. The problem addressed is the inefficiency of direct tunnels between servers hosting VMs, which consumes resources on both servers. The solution involves a storage medium containing instructions for managing communication between a first VM in a first server and a second VM in a second server. A network switch and a second switch establish a tunnel between them, enabling the VMs to communicate without requiring a direct tunnel between the servers. This tunnel is transparent to the servers, meaning they are unaware of its existence. By avoiding direct server-to-server tunnels, the solution reduces resource utilization on both servers, improving efficiency. The tunnel is managed by the switches, which handle the communication overhead, allowing the servers to focus on their primary tasks. This approach is particularly useful in cloud computing and data center environments where multiple VMs across different servers need to communicate securely and efficiently.
Unknown
January 2, 2018
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.